INDEX

Explanations

references to utility classes and core libraries in programming

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

HostException

-0.47

DataPropertyName

-0.47

LLocation

-0.45

 EconPapers

-0.43

 TestBed

-0.42

 wireType

-0.42

ModelSerializer

-0.39

abstractmethod

-0.38

findpost

-0.38

RegressionTest

-0.38

POSITIVE LOGITS

util

1.16

UTIL

0.80

 util

0.79

utili

0.59

Util

0.58

 pinulongan

0.58

utils

0.58

 Util

0.57

 utilization

0.57

 FormData

0.56

Activations Density 0.002%

references to utility classes and core libraries in programming

No Comments

No Known Activations